February 2026 - Wine-gitlab

[PATCH 0/4] MR10104: user32/edit: Reset capture on WM_LBUTTONDOWN if already enabled
by Ivan Ivlev (＠iviv) 03 Mar '26

03 Mar '26

Edit control does not release mouse capture after WM_LBUTTONDOWN has been processed two times in a row. This MR fixes this issue by releasing capture and setting it again on WM_LBUTTONDOWN if window was already captured. I have stumbled upon the app in which edit control process WM_LBUTTONDOWN twice on one mouse click. While in windows it causes no problem, wine edit does not release capture, so other buttons inside this app become unclickable. It happens because when WM_LBUTTONDOWN is processed twice, WM_CAPTURECHANGED is sent and bCaptureState is set to FALSE, then EDIT_WM_LButtonUp does not release capture because bCaptureState is already FALSE. Thing about tests: I have added test case for this problem to be visible. I also had to use todo_wine because SendMessageA for WM_LBUTTONDOWN and WM_LBUTTONUP returns 0 in wine and 1 in windows (which is not the case for combobox test, where SendMessageА for WM_LBUTTONDOWN also returns 1 in wine). -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10104

4 7

[PATCH 0/1] MR10059: comctl32: Fix the bug when the cInitial is too large which lead to...
by Kun Yang (＠yangkun) 03 Mar '26

03 Mar '26

I'm sorry I have pushed another mr for this topic due to my fault for pushing to the master branch of my forked project. This is about the failure of CreateDIBSection in ImageList_Create. Last discussion is in !10039 (https://gitlab.winehq.org/wine/wine/-/merge_requests/10039) I have made my last fix in https://gitlab.winehq.org/wine/wine/-/merge_requests/10039#note_129134. Please review it again @zhiyi . I am sorry for making a new merge request again. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10059

3 2

[PATCH 0/1] MR10213: winmm/tests: Reserve additional memory in test_formats (ASan).
by Bernhard Übelacker (＠bernhardu) 03 Mar '26

03 Mar '26

Followup of fe239731. In `test_formats` only `sizeof(WAVEFORMATEXTENSIBLE)` gets allocated, but in `WINMM_OpenDevice` it tries to read one addional byte in the "cbSize += 1" test. CC: @giomasce [Test pattern page](https://test.winehq.org/data/patterns.html#winmm:wave) (This patch won't help here.) [Testbot run with this patch](https://testbot.winehq.org/JobDetails.pl?Key=162097) (Shows also the other issues like the test pattern page.) <details> <summary>ASan details [gitlab ci run](https://gitlab.winehq.org/bernhardu/wine/-/jobs/232675#L2501)</summary> ``` ==winmm_test.exe==320==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x0022fc78 at pc 0x784c4532 bp 0x04d3fbec sp 0x04d3f7bc READ of size 41 at 0x0022fc78 thread T243 #0 0x784c4531 in __asan_memcpy /home/runner/work/llvm-mingw/llvm-mingw/llvm-project/compiler-rt/lib/asan/asan_interceptors_memintrinsics.cpp:65:3 #1 0x78f3210e in WINMM_OpenDevice /builds/bernhardu/wine/build64/../dlls/winmm/waveform.c:1124:9 #2 0x78f30aad in WOD_Open /builds/bernhardu/wine/build64/../dlls/winmm/waveform.c:1277:11 #3 0x78f3638e in WINMM_DevicesMsgProc /builds/bernhardu/wine/build64/../dlls/winmm/waveform.c:2393:16 #4 0x7b13b573 in WINPROC_wrapper (C:\windows\system32\user32.dll+0x1004b573) #5 0x7b13bdf3 in call_window_proc /builds/bernhardu/wine/build64/../dlls/user32/winproc.c:111:15 #6 0x7b13beea in dispatch_win_proc_params /builds/bernhardu/wine/build64/../dlls/user32/winproc.c #7 0x7b13c807 in User32CallWinProc /builds/bernhardu/wine/build64/../dlls/user32/winproc.c:827:14 #8 0x7be1dc86 in dispatch_user_callback /builds/bernhardu/wine/build64/../dlls/ntdll/exception.c:297:18 #9 0x7be46a62 in KiUserCallbackDispatcher /builds/bernhardu/wine/build64/../dlls/ntdll/signal_i386.c:205:23 #10 0x79d81d3b in NtUserPeekMessage (C:\windows\system32\win32u.dll+0x10011d3b) #11 0x78f361d6 in WINMM_DevicesThreadProc /builds/bernhardu/wine/build64/../dlls/winmm/waveform.c:2475:16 #12 0x784d6f21 in asan_thread_start(void*) /home/runner/work/llvm-mingw/llvm-mingw/llvm-project/compiler-rt/lib/asan/asan_win.cpp:147:14 #13 0x7be4678a in call_thread_func_wrapper (C:\windows\system32\ntdll.dll+0x7bc4678a) #14 0x7be470fa in call_thread_func /builds/bernhardu/wine/build64/../dlls/ntdll/signal_i386.c:503:9 Address 0x0022fc78 is located in stack of thread T0 at offset 56 in frame #0 0x0043676f in test_formats /builds/bernhardu/wine/build64/../dlls/winmm/tests/wave.c:2434 This frame has 1 object(s): [16, 56) 'fmt' (line 2455) <== Memory access at offset 56 overflows this variable HINT: this may be a false positive if your program uses some custom stack unwind mechanism, swapcontext or vfork (longjmp, SEH and C++ exceptions *are* supported) SUMMARY: AddressSanitizer: stack-buffer-overflow /builds/bernhardu/wine/build64/../dlls/winmm/waveform.c:1124:9 in WINMM_OpenDevice ``` </details> -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10213

4 4

[PATCH v2 0/6] MR9348: comctl32/imagelist: Pre-multiply rgb by the alpha channel in add_dib_bits for v6.
by Ziqing Hui (＠zhui) 03 Mar '26

03 Mar '26

We used to do this when drawing the image. But our tests show that for v6, the image bitmaps have been alpha pre-multiplied before drawing. So we should do this when adding images. -- v2: comctl32/imagelist: Support adding 32bpp images to a non-32bpp image list for v6. comctl32/imagelist: Do alpha blending when adding 32bpp images to a non-32bpp image list. comctl32/imagelist: Introduce create_dib_section helper. comctl32/imagelist: Introduce image_list_color_flag helper. comctl32/imagelist: Pre-multiply rgb by the alpha channel in add_dib_bits for v6. comctl32/imagelist: Introduce premultiply_alpha_channel helper. https://gitlab.winehq.org/wine/wine/-/merge_requests/9348

3 8

[PATCH 0/1] MR10201: dwrite: Add locales to locale list.
by Daniel Lehman (＠dlehman25) 03 Mar '26

03 Mar '26

The Chinese version of the Noto Sans CJK font was being chosen despite being in Japanese locale. Specifically, Noto Sans CJK SC was used for 0x5c06 instead of Noto Sans CJK JP. The locale list for the system fallback collection was just the neutral locale. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10201

3 5

[PATCH 0/1] MR10206: comctl32/treeview: Fixed missing redraw of item being unfocused
by Piotr Pawłowski (＠DEATH) 03 Mar '26

03 Mar '26

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/10206

2 2

[PATCH 0/1] MR10205: comctl32/treeview: Fixed wrong check for currently edited item
by Piotr Pawłowski (＠DEATH) 03 Mar '26

03 Mar '26

Wine-Bug: https://bugs.winehq.org/show_bug.cgi?id=59455 -- https://gitlab.winehq.org/wine/wine/-/merge_requests/10205

3 4

[PATCH 0/1] MR10207: comctl32/treeview: Cycle all state images from custom image list when toggling checkboxes
by Piotr Pawłowski (＠DEATH) 03 Mar '26

03 Mar '26

-- https://gitlab.winehq.org/wine/wine/-/merge_requests/10207

2 3

[PATCH v3 0/2] MR10008: start: Fix arg parsing for quoted paths ending with backslash.
by Yeshun Ye (＠yeyeshun) 02 Mar '26

02 Mar '26

When using the `start` command with a `/d` switch specifying a quoted path that ends with a backslash (e.g., `start /d "c:\path\" app.exe`), subsequent arguments were incorrectly concatenated into the path string due to parsing ambiguity. The error I encountered was with `start /d "c:\temp\" test.exe`, which was parsed into the arguments `/d` and `"c:\temp\" test.exe`. The path and the executable were merged into a single argument as a new path, preventing the child process from launching. In my first patch, I attempted to identify and split this combined argument, but this caused test failures in `kernel32`. I then retested on Windows and found that Windows also parses it as `/d` and `"c:\temp\" test.exe`—Wine's parsing is consistent with Windows. However, `start` works correctly on Windows because it is a built-in command, whereas in Wine, `start` is an external executable, leading to different behavior. Signed-off-by: YeshunYe yeyeshun(a)uniontech.com -- v3: start: Re-parsing the command line. https://gitlab.winehq.org/wine/wine/-/merge_requests/10008

3 5

[PATCH 0/1] MR7419: wininet: Fix crash when connection is null in NETCON_secure_connect
by Yongjie Yao (＠yaoyongjie) 02 Mar '26

02 Mar '26

There will be crashed when connect is null, so we need avoid crash by checking connect value. -- https://gitlab.winehq.org/wine/wine/-/merge_requests/7419

4 3